Long-term heart preservation by intermittent perfusion with crystalloid medium

Long-term heart preservation by intermittent perfusion with crystalloid medium This study was undertaken to determine whether hearts preserved with intermittent coronary perfusion would recover physiologic function after a prolonged period of hypothermic preservation. Intermittent perfusion is commonly used for cardioplegia, but its efficacy in long-term heart preservation has not yet been demonstrated. Five groups of isolated rat hearts were studied (n = 7 per group): (1) fresh nonpreserved control hearts; (2) hearts preserved with continuous low-pressure perfusion via the aorta; (3) hearts preserved with cycles of 5 minutes of perfusion foUowedby 25 minutes of nonperfusion; (4) hearts preserved with cycles of 10 minutes of perfusion foUowed by 25 minutes of nonperfusion; (5) hearts preserved with submersion storage without perfusion. An oxygenated extraceUular-type crystalloid medium (oxygen tension = 820 ± 5 mm Hg) was used as a preservation medium; preservation was for 12 hours. During preservation, the coronary resistance of the intermittent perfusion-preserved hearts increased significantly, and these hearts produced significantly more excess lactate than did hearts in the other two preservation groups. The submersion-stored hearts exhibited no postpreservation ventricular function in an isolated perfused working rat heart system. The poststorage function of the other four groups, which was quantified during a 4-hour, 37 0 C perfusion period at constant heart rate, indicated that there were no significant group differences with respect to output or energetics (coronary flow, aortic output, cardiac output, myocardial oxygen consumption, and external work efficiency). The intermittent perfusion-preserved hearts had significantly lower postpreservation contractile function (left ventricular systolic pressure, peak rates of left ventricular pressure development and relaxation, peak aortic flow rate, stroke work, and peak power) and higher left ventricular end-diastolic pressure compared with the control group. Although hearts preserved with intermittent perfusion had a loss of contractile function and decreased compliance compared with fresh hearts, after preservation they had better function than did hearts preserved with submersion storage and the same function as hearts preserved with continuous perfusion. (J THoRAc CARDIOVASC SURG 1993;106: 811-22)

Leigh D. Segel, Phl)," and David M. Follette, MD,b Davis and Sacramento, Calif.

A

number of issues must be addressed in determining strategies for long-term heart preservation in a clinical setting. One issue is the method of organ storage. The From theDepartments of Internal Medicine" and Surgery," University of California School of Medicine, Davis, Calif., and University of California, Davis, Medical Center, Sacramento, Calif. Supported by grant No. ROI HL30065 from the National Institutes of Health. Received for publication Aug. 27, 1992. Accepted for publication Nov. 12, 1992. Address for reprints: Leigh D. Segel, PhD, Division of Cardiovascular Medicine, TB 172, University of California School of Medicine, Davis, CA 95616. Copyright © 1993 by Mosby-Year Book, Inc. 0022-5223/93 $1.00 +.10

12/1/44499

simplest method is the one currently used in clinical practice: submersion of the heart in a cold crystalloid solution. However, this does not appear to be the optimum procedure for prolonged. storage times. Several studies have suggested that continuous perfusion through the coronary vasculature during preservation produces better maintenance of adenosine triphosphate stores! and better recovery of cardiac function-" than does submersion storage. But prolonged in vitro perfusion with aqueous media may have adverse complications, including edema. Preservation with intermittent perfusiorrv" might be advantageous from a practical standpoint, compared with continuous perfusion, if postpreservation function and substructural integrity are comparable. However, even in brief cardioplegia scenarios, tissue damage can result

8II

8 12

The Journal of Thoracic and Cardiovascular Surgery November 1993

Segel and Follette

120

~

100

~

CJ)

80

E E

60

::c

40 20

-

-

0 5 4

E

01

3

E E

2 f-

.

~

8

~

6

E

4

Ql

.., 0

-

" Ql fIl

~

I

-

150

CJ)

U.

'3

100

E

50

Ql

.., 0

0

I

I

-.e-e-e-'iii 'i ~ i

I

I

--

F4

dP/dtmax I

!-1

I

I

I

I

I

I

T

~ 1&"

I

0 200

I

FFi=Fi

l.

l.

::r 1.

I

..J.

1.

1

1

r'r

1

P-

Stroke Work

2

I"

,

-----. pe I

I

0 10

'3

,

8

~

~

CJ)

LVSP

IIIii

fIl

::c

;=;-;

~

,..

I

I

I

I

I

R~ -==J=----F ~-t=--R Power max I

I

I

I

I

I

I

0.0

0.5

1.0

1.5

2.0

2.5

3.0

, 3.5

4.0

Hours of Paced Working Heart Perfusion Fig. 1. Contractile function indexes of nonpreserved and preserved isolated working rat hearts during 4 hours of paced antegrade perfusion. e, Control group; 0, continuous perfusion group; .A, intermittent group II; £::", intermittent group I; Powermax , peak power. See text for explanation of treatment groups.

from intermittent perfusion. Recently, von Oppell and associates? reported that postischemic release of lactate dehydrogenase was greater in isolated rat hearts that underwent multidose cardioplegia than in those that

underwent single-dose cardioplegia, even though the multidose hearts had better functional recovery. Repeated bouts of ischemia can result in loss of adenine nucleotide precursors.v 9 which could preclude functional

The Journal of Thoracic and Cardiovascular Surgery Volume 106, Number 5

Segel and Follette 8 I 3

Table I. Coronary flow and arterial - venous oxygen tension difference at start and end ofpreservation Coronary flow [mlfmin . gm dry weight) Group

Start (0 hr)

End (12 hr)

Continuous perfusion IntermittentI Intermittent II

12.0 ± 0,5A 12.2 ± 0.4A 12.6 ± 0.5A

11.0 ± 0.8A 7.7 ± 0.8A 9.7 ± O.sA,B

Pao2*-Pvo2 (mmHg) Ohrvs.12hr

NSD p
p
Start (0 hr)

End (12 hr)t

ohr vs. 12 hr

298 ± 2QA 292 ± 15A 287 ± 14A

331 ± 25A 452 ± 2()B 402 ± 13B

p
Values are mean ± standard error of the mean. In each column, values that differ significantly have no letters in common (one-way analysis of variance). The Start versus End comparisons were done by two-tailed paired t tests. NSD, Not significantly different (p> 0.05). *Pao2 = 820 ± 5 mm Hg. tPv02 was measured 10 minutes after starting the last flow cycle in the intermittent groups.

Table II. Excess lactate production duringpreservation First 30 ml of 37° C coronary reperfusate (umol lactatefgm dry weight)

Preservation medium (umol lactatefl l hr . gm dry weight)

Group

49 ± 291 ± 278 ± 52 ±

Continuous perfusion Intermittent I Intermittent II No perfusion

8± 19 ± 24 ± 72 ±

lQA

13B

13B

6A

Preservation medium and reperfusate [umol lactatetgm dry weight)

IA

57 ± lQA 31O±17 B 301 ± 17B 125 ± 9c

6A

4A 4B

Values are mean ± standard error of the mean. In each column, values that differ significantly have no letters in common (one-way analysis of variance).

Table III. Coronary flow and energetics ofhearts after 30 minutes of aorticretrograde reperfusion at 37° C Coronary flow (mlfmin , gm dry weight)

Group

Control Continuous perfusion Intermittent I Intermittent II

86.4 ± 65.2 ± 58.2 ± 58.2 ±

8.7A 4.6A,B 3.9B 3.9B

MV02 (ml oxygenfmin . gm dry weight) 0.350 ± O.027A

Oxygen supply/use [ml oxygenimin . gm + ml oxygen/min. gm)

4.63 ± 3.23 ± 2.19 ± 2.05 ±

0.404 ± 0.043A. B 0.502 ± 0.031B, C 0.538 ± 0.027c

0.67A 0.63A,B 0.29B O.l9B

Pv02 (mmHg)

438 ± 349 ± 283 ± 280 ±

21A 27A,B 28B 22B

Valuesare mean ± standard error of the mean. In each column, values that differ significantly have no letters in common (one-way analysis of variance). MV02, Myocardial oxygen consumption.

Table IV. Function of hearts during nonpaced workingheart reperfusion Group

Control Continuous perfusion Intermittent I Intermittent II

LVSP (mmHg)

dP/dt mox (mm Hg/msec)

128 ± 4A 112 ± 4 B 103 ± 3B

5.14 ± O.l6A 4.39 ± 0.34A , B 4.13 ± O.l9B 3.80 ± O.l5B

102

± 4B

LVEDP (mmHg)

5.3 ± 5.9 ± 9.6 ± 9.6 ±

1.3A

0.8A, B 1.3B

0.8B

Stroke work (mloulesfgm}

14.1 ± 11.2 ± 9.1 ± 9.0 ±

1.4A 1.4A 1.4A 1.3A

Peak power

(mlouleslgm . sec) 283 ± 223 ± 174 ± 174 ±

22A

25A,B 25B 24B

Coronary flow (mlfmin . gm)

90.1 ± 75.8 ± 64.1 ± 60.4 ±

3.6A 3.7A,B 4.8B, C

2.6c

Cardiac output

(mlfmin . gm) 393 ± 341 ± 282 ± 296 ±

24A 14A,B 29B 19B

Values are mean ± standard error of the mean. In each column, values that differ significantly have no letters in common (one-way analysis of variance).

recovery. In the experiments of Flameng, Dyszkiewicz, and Minten," intermittently perfused dog hearts, unlike continuously perfused hearts, had no recovery of function after 24 hours of preservation at 0.5° C. We undertook the present study to determine whether intermittent perfusion could be successfully used for prolonged heart preservation.

Materials and methods Animals and heart preservation methods. Procedures and animalcare conformedto the "Principlesof Laboratory Animal Care" formulated by the National Society for Medical Research and the "Guide for Care and Use of Laboratory Animals" prepared by the Institute of Laboratory Animal Resources and published by the National Institutes of Health (NIH PublicationNo. 86-23,revised 1985).All animal use and

The Journal of Thoracic and Cardiovascular Surgery November 1993

8 I 4 Segel and Follette

Table V. Average values of cardiac function indexes during 4 hours of working heart perfusion Group Control Continuous perfusion Intermittent I Intermittent II Pooled SD

LVSP \14 ± IA lOS ± 2A. B 99 ± 4B 100 ± 2B 6

dP/dtmax 0.08A O.lOB O.l6B O.l4B

4.S9 ± 4.10 ± 3.93 ± 3.94 ± 0.32

-dP/dtmax 3.37 ± O.lSA 2.86 ± 0.09A• B 2.59 ± 0.28B 2.S4 ± 0.09B 0.4S

AF ratemax 221 ± 4A 204 ± 7A. B 180 ± l3 B 188 ± 7B

22

Stroke work 0.58A 0.58A. B 0.8SB 0.64A. B

9.82 ± 8.49 ± 7.42 ± 7.88 ± 1.32

Peak power 206 ± 13A 170 ± \lA.B lSI ± 18B IS6 ± 14B 28

LVEDP 3.8 ± 0.2A S.O ± O.4A. B 8.1 ± I.SB 7.6 ± 0.8B 2.3

Values are mean ± standard error of the mean. In each column, groups that differ significantly have no letters in common. The pooled standard deviations were used to determine group differences in the repeated-measures analyses of variance. -dP/dt max, Peak rate of left ventricular relaxation (mm Hg/rnsec); AFratemax , peak aortic flow rate (ml/rnin . gm dry weight); AO, aortic output (rnl/min . gm dry weight); EFF, external work efficiency (0/0); LAC/SW. lactate production rate/stroke work (mmol lactate/D.S hr . Joule); SD, standard deviation. For other units, see Tables III and IV.

care procedures were approved by the Animal Use and Care Administrative Advisory Committee of the University of California at Davis. Male Sprague-Dawley rats (379 ± 7 gm; Charles River Laboratories, Inc., Wilmington, Mass.) were housed individually and given free access to Purina Rodent Laboratory Chow (Purina Mills, St. Louis, Mo.) and water. Hearts were removed from heparinized rats (1000 U heparin sodium/kg body weight, administered intraperitoneally) under anesthesia (acepromazine maleate 0.75 rng/kg body weight + xylazine 6 mg/kg + ketamine 60 mg/kg, administered intramuscularly). Rats' lungs were ventilated with room air during cardiectomy. The heart was immersed in 23 0 C cardioplegic solution (described later), and the ascending aorta was ligated to a flanged cannula (PE240; Becton-Dickinson, Parsippany, N.J.). Coronary perfusion via the aortic root at 93 mm Hg pressure was begun within 5 minutes after excision. Temperature of the coronary perfusate was gradually lowered from 23 0 C to 130 C by perfusing the heart first with room temperature cardioplegic solution and then with iced cardioplegic solution (approximately 40 ml for 3 minutes). The heart was immediately transferred to the preservation apparatus as previouslydescribed, 10 with one modification: a solenoid pinch clamp (two-way NC No. 225POI3-2Ia; Neptune Research, Maplewood, N.J.) controlled by a time-delay relay (No. SSAC TDR4A33; Steven Engineering, South San Francisco, Calif.) was placed around the aortic inflowtubing. The relay repeatedly opened and closed the pinch clamp throughout the 12-hour preservation period, thus allowing intermittent inflow of preservation fluid through the tubing into the aorta and the coronary vasculature. Hearts were preserved for 12 hours at 120 C and 18 mm Hg perfusion pressure with either continuous perfusion or one of two intermittent perfusion protocols. In one intermittent protocol, the cycle included 5 minutes of perfusion followed by 25 minutes of nonperfusion. In the other intermittent protocol, the cycle included 10 minutes of perfusion followedby 25 minutes of nonperfusion. The arterial oxygen tension of preservation fluid entering the heart from the reservoir (Pao2) and the oxygen tension of the coronary venous effluent (Pvo-) were measured at the start and end of the preservation period. Hearts that were preserved with submersion storage were excised and arrested with cardioplegic solution as described previously and then were perfused for 3 minutes with preservation solution before being submerged in that medium at the same temperature and Pao- used for the other preserved hearts.

Cardioplegia and preservation media. Solutions were made with Milli-Q Reagent Grade water (Millipore Corp., Bedford, Mass.), oxygenated with 100% oxygen, and then vacuumfiltered (0.8!Lm porosity Millipore filter). The cardioplegic and preservation media were slightly modified versions of those of Wicomb and associates. I I The oxygen tension (P02) and pH values were measured on a Radiometer ABL3 acid-base analyzer (Radiometer A/S, Copenhagen, Denmark) at 370 C. The cardioplegic solution contained the following, in millimoles per liter: NaCI, 102; NaHC0 3, 4; KCI, 10; CaCb, 1.1; MgS04, 14; glucose, 278; and procaine HCI, 0.46. Theconcentration of heparin sodium used was 1000 U /L. The pH was 7.58 ± 0.02, and P02 was 384 ± 12 mm Hg. The preservation solution contained the following, in millimoles perliter: NaCI, 144;CaCh, 1.1;KH 2P04, 1.74;K2HP04, 6.35; MgS04, 14; glucose, 11.1; sucrose, 7; glycerol, 136; taurine, 4; and procaine HCI, I. The concentration of heparin sodium used was 100 U /L. The solution was equilibrated with 100% oxygen throughout the preservation period to give a P0 2 of 820 ± 5 mm Hg and a pH of 6.88 ± 0.01. After 12 hours, hearts that had been intermittently perfused underwent a final 12 minutes of perfusion, at which time Pv02 was measured, before removal from the apparatus. Measurement of cardiac function with isolated heart perfusion. After preservation, hearts were placed in 230 C KrebsHenseleit.medium (composition described later) for 2 minutes and then were connected to the isolated heart perfusion apparatus previously described.l- 13 Coronary perfusion at 37 0 C and 60 mm Hg via the aorta was immediately begun. The heart Was reperfused retrogradely (nonworking or Langendorff method) via the aorta while the left ventricle and left atrium cannulations were done; cardiac function was then recorded. The heart was then reperfused antegradely (working), with the left ventricle fillingvia the left atrium and ejecting via the aorta. The working hearts were equilibrated at spontaneous rates for 10 minutes, after which function was recorded. Electrical pacing via the right atrium was then begun (3 V, 5.5 msec) to give a heart rate of 322 beats/min. Function was recorded after a 5-minute stabilization period while the heart was paced (0 hour) and every 30 minutes thereafter for 4 hours. After the 4-hour measurement point, the {J-adrenergicagonist dobutamine (0.1 !Lmol/L final concentration) was added to the perfusate,and a final measurement of function was made. The dobutamine test was used to evaluate cardiac responsiveness to increased demand. At the end of the experiment, the left and right ven-

The Journal of Thoracic and Cardiovascular Surgery Volume 106, Number 5

Coronary flow 5.8A

90.2 ± 79.1 ± 2.9A 72.5 ± 9.4A 69.2 ± 2.6A 14.2

AO 309 283 260 276

16A

± ± 15A ± 29A ± 21A 38

Segel and Follette

Cardiac output 399 362 333 346

17A

± ± 16A ± 37 A ± 23 A 51

MV02 1.31 1.21 1.18 1.11

0.08 A

± ± 0.04 A ± 0.14 A ± 0.03 A 0.21

tricleswere immediately weighed to obtain postperfusion wet weights before being dried at 110° C for 16 hours. The following cardiac function indexes were measured and computed as previously described 10, 13: left ventricular peak systolic pressure (L VSP), peak rates of left ventricular pressure development (dP /dtmax ) and relaxation (-dP/dtmax) , left ventricular end-diastolic pressure (LVEDP), peak aortic flow rate, aorticoutput, coronary artery flow, total cardiac output, stroke work, and peak power. Arterial and coronary venous P02 values (Pao-, Pv02) were obtained without exposing the perfusate samples to air. Myocardial oxygen consumption and external work efficiency were computed lO, 13 from the measured indexes. The oxygen supply/use ratio was calculated as Coronary flow X Arterial perfusate oxygen content/Myocardial oxygen consumption. The height of the fluid column entering the left atrium during the working heart perfusion was 12 cm; the hydrostatic afterload produced a mean perfusion pressure of 58 mm Hg. At each measurement point, a sample of the recirculating perfusate was taken for lactate assay." The reservoir was replenished so that the perfusate volume remained at 250 ml, and the dilutions were taken into account when calculating lactate production rates. The Krebs-Henseleit perfusate had the followingcomposition (in millimoles per liter): NaCl, 125; KCI, 4.9; CaCh, 2.8; MgS04, 1.36; KH 2P04, 1.36; NaHC0 3, 20.1; glucose, 11.1; and Na2 ethylenediaminetetraacetate, 0.45. The perfusate, prepared with Milli-Q Reagent Grade water, was equilibrated with 95% oxygen/5% carbon dioxide and filtered (0.8 !Lm porosityMillipore filter). During the experiment, the perfusate wascontinuously equilibrated with 95% oxygen and 5% carbon dioxideto produce a pH of 7.41 ± 0.01 and a P02 of 60 I ± 2 mmHg. Data analysis. Seven hearts per group were preserved with oneof the four preservation protocols previously described: continuous perfusion; intermittent group I, perfusion for 5 minutes and nonperfusion for 25 minutes; intermittent group II, perfusion for 10 minutes and nonperfusion for 25 minutes; and no perfusion.The hearts were then studied in the isolated working heart system to quantify function. Seven nonpreserved fresh hearts that were studied as isolated working hearts to determine their function served as controls. Data are reported as mean ± standard error of the mean. Statistical analyses were conducted with Statistical Analysis Systems software (SAS Institute, Cary, N.C.). Cardiac function data obtained during the working heart period (0 hour through 4 hours of paced function) were analyzed with repeated-measures analysis of covariance with dry heart weight as a

EFF 11.8 10.6 9.7 10.8

Pv02 A

± l.l ± 0.6 A ± 0.8 A ± 0.7 A 1.9

143 124 94 100

15A

± ± llA.B ± 5B ± 9B 26

Oxygen supply/ use 1.33 1.26 1.19 1.21

A

± 0.05 ± 0.03 A• B ± O.OIB ± O.Q2B 0.08

8 15

LAC/SW 3.1 5.2 18.8 13.3

± 0.5 A ± LoA ± 7.0 B ± 3.3 A• B 9.1

covariate. This analysis was followed by an orthogonal polynomial decomposition; the coefficients of the orthonormal polynomials were calculated with the Gram-Schmidt method. The Huynh-Feldt adjustment was used in the univariate tests for within-subject effects. For other comparisons of the groups, a one-way analysis of variance followed by Tukey's honest significant difference method was used. Statistical significance was accepted at p < 0.05.

Results Preservation period. Coronary flow in intermittent groups I and II declined significantly during the 12 hours of preservation, indicating an increase in coronary resistance, whereas that of the hearts with continuous perfusion remained stable (Table I). At the start of preservation, those three groups had the same Pao- - Pvo , values; at the end, intermittent groups I and II had lower Pvovalues, and consequently greater Paoz - Pvo, values, than did the continuous perfusion group (Table I). We measured the excesslactate produced by the hearts that appeared in the preservation medium and the lactate that appeared in the first 30 ml of coronary effluent on 37° C reperfusion (Table II). Significantly more lactate was produced by hearts in intermittent groups I and II than by hearts in the continuous group. Only a small amount of lactate appeared in the preservation medium of the hearts that underwent no perfusion because there was no perfusion through the coronary vasculature during preservation. During 37° C reperfusion, however, a relatively large amount of lactate was flushed from the nonperfused hearts. Function of submersion-stored hearts. All of the hearts that underwent no perfusion exhibited contracture after preservation and had no function when reperfused at 37° C. The following results therefore do not include data from this group. Nonworking (Langendorff) function of the hearts. All of the hearts in the control, continuous perfusion, and both intermittent groups resumed spontaneous contractions within seconds of reperfusion at 37° C. After 30 minutes of Langendorff perfusion, there were no group

The Journal of Thoracic and Cardiovascular Surgery November 1993

8 1 6 Segel and Follette

I

C)

I

100 80

C

60

E

40

'E

20

Coronary Flow

0 300 I

C)

250

I'c

200

E

150

'E

100 50

I

I

C)

c

'E

E

Aortic Output

0 400 350 300 250 200 150 100 50 0

Cardiac Output 0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

Hours of Paced Working Heart Perfusion Fig. 2. Output and flow indexes of nonpreserved and preserved isolated working rat hearts during 4 hours of paced antegrade perfusion.•, Controlgroup; 0, continuous perfusion group; ... , intermittentgroup II; L, intermittent group 1.

differences in function (LVSP, dP /dtmax, -dP/dtmax, and LVEDP) but there were differences in coronary flow and energetics (Table III). Coronary flow values of the hearts of both intermittent groups were significantly lower than those of control hearts. The PV02 values were parallel to the coronary flow values, and the oxygen supply/use ratio indicated that control hearts had an abundant oxygen supply compared with hearts in both intermittent groups (Table III). Working (ejecting) function of the hearts. All of the hearts in the control, continuous perfusion, and both

intermittent groups began ejecting when ventricular filling via the left atrium was initiated. Spontaneous heart rates of the four groups were not different, averaging 264 ± 9 beats/min. However, there were significant differences in contractile indexes (LVSP, dP /dtmax, -dP/ dt max , LVEDP, and peak power) and output (coronary artery flow, aortic output, and total cardiac output) at this point (Table IV; -dP/ dt max and aortic output not shown). The hearts of both intermittent groups exhibited lower function than that of control hearts. The continuous perfusion group had intermediate function that was not sig-

The Journal of Thoracic and Cardiovascular Surgery

Segel and Follette

Volume 106, Number 5

8I7

1.6 1.4

1.2 Cl I'

e

'E o

'N

E

1.0

0.8 0.6 0.4

0.2 0.0 ......--I._---I._---l"------'L..-_L.-_........_..&.-_..L-_....&..---i

220 200 180 Cl

:I:

E E

160 140

120 100 80 60 40

I

Cl

20 0 ......- ' - -........-

........-

........-

........---1.---1.---1.----&.--1

1.6

I

1.4

o E

N

~-I

Cl

I'

c

'E 'N

o E

1.2 1.0

0.8 0.6

0.4

02 Supply/02 Utilization

0.2 0.0 1-......._ ........_ ........_--&._--&._--&._--&._--'-_--'-_-1 12 10

8 o~

6 4 2

Efficiency

01...--1.._-1.._........_

0.0

0.5

1.0

........_

1,5

........_ --'-_--'-_--'-_---&._.....

2.0

2.5

3.0

3.5

4.0

Hours of Paced Working Heart Perfusion Fig. 3. Myocardial energetics of nonpreserved and preserved isolated working rat hearts during 4 hours of paced antegrade perfusion.•, Control group; 0, continuous perfusion group; . , intermittent group II; b., intermittent group I; MV02, myocardial oxygen consumption.

nificantly different from the control group or the intermittent groups, except for LVSP, which was lower than thatof the control group. All four groups exhibited statistically similar stroke work values. Data obtained during the 4-hour paced working peri-

od are shown in Figs. I to 5 (-dP/ dt max and peak aortic flow rate data not shown) and Table V. During pacedheart perfusion, significant group differences were apparent in the contractile indexes but not in the output indexes. Overall, for the 4-hour period, the hearts of both

8I8

The Journal of Thoracic and Cardiovascular Surgery November 1993

Segel and Follette

10 8 C)

:c

E E

6 4 2 0

m---f=4=t-~ •

•

•

•

•

I

1.0

1.5

2.0

2.5

LVEDP 0.0

0.5

3.0

3.5

4.0

Hours of Paced Working Heart Perfusion Fig. 4. LVEDP of nonpreserved and preserved isolatedworking rat hearts during 4 hours of paced antegrade perfusion. e, Control group; 0, continuous perfusion group; "', intermittent group II; c; intermittent group I. intermittent groups had significantly lower LVSP, dPI dtmax, -dP I dtmax, peak aortic flow rate, and peak power and significantly higher LVEDP than that of control hearts (Table V). With respect to stroke work, intermittent group I but not intermittent group II differed significantly from the control group. No significant differences in function were detected among the continuous perfusion and intermittent groups. The continuous perfusion group differed significantly from the control group only in dP Idtmax, which was lower in the continuous perfusion group. No significant differences were detected among the four groups in coronary artery flow, aortic output, total cardiac output, myocardial oxygen consumption, or external work efficiency. Analysis of the first-degree polynomial fit of the perfusion data for the 4-hour period indicated that the control group had higher initial values and declined more than the other groups in several indexes (Figs. 1 and 2). This decline was significant for LVSP (control versus continuous perfusion), dPI dtmax (control versus intermittent group I and continuous perfusion), and coronary flow (control versus both intermittent groups). The seconddegree polynomial fit indicated that the values of the continuous perfusion group and both intermittent groups increased during the early part of perfusion (up to the l-hour point) and then subsequently declined (at 3 to 4 hours). A significant negative curvature of LSVP, peak aortic flow rate, stroke work, and peak power data was detected for the continuous perfusion and intermittent groups; another such curvature was detected for dP I dtmax, coronary artery flow, and total cardiac output for both intermittent groups. The Pv02 and oxygen supplyluse ratio of control hearts during perfusion were, on the average, higher than those values for the hearts of both intermittent groups (Table

V; Fig. 3). The first-degree polynomial fit of those data indicated that control heart Pv02 and oxygen supplyluse ratio declined faster than those values of the other three groups, and the values for continuous perfusion hearts declined faster than those values of intermittent group II hearts. Excess lactate production during the 4-hour perfusion period, normalized to stroke work, was lower in the control and continuous perfusion groups than in intermittent group I (Table V; Fig. 5). Response of the hearts to dobutamine. All of the hearts of the control, continuous perfusion, and both intermittent groups exhibited an inotropic response to dobutamine at the end of the 4-hour paced perfusion period. The same pattern offunction differences observed during the 4-hour period was also apparent at this point. Contractile indexes (LVSP, dP/dtmax) of the hearts of both intermittent groups were lower than those of the control hearts (Table VI). The LVEDP value of intermittent group I was higher than that of control (Table VI). Values of the other function indexes showed no significant group differences. Heart weight and fluid content. As shown in Table VII, the hearts of the control, continuous perfusion, and both intermittent groups had higher amounts of tissue fluid after the 37° C Krebs-Henseleit perfusion than did fresh nonperfused rat hearts, which we determined to be 3.37 ± 0.03 ml water/gm dry weight tissue.!? The intermittent group I hearts had significantly more edema at the end of perfusion than did control hearts (Table VII). Discussion M ultidose, intermittent cardioplegia for adult hearts is widely used clinically and has been shown to be advantageous compared with single-dose cardioplegia.I: 15, 16 The

The Journal of Thoracic and Cardiovascular Surgery Volume 106, Number 5

Segel and Follette

Table VI. Function of isolated hearts during dobutamine treatment Group

Control Continuous perfusion Intermittent I Intermittent II

LVSP (mmHg)

139 ± 122 ± 116 ± 117 ±

3A 3A, B 7B 5B

dP/dt max (mm Hgfmsec}

6.26 ± 5.46 ± 5.24 ± 5.07 ±

0.17A 0.17A, B 0.41B O.l8B

8 19

Table VII. Dry weight and fluid content of hearts after 37° C perfusion LVEDP (mmHg)

Group

4.2 ± 0.3A 5.6 ± 0.7A, B 8.9 ± 2.0B 8.3 ± l.lA,B

Control Continuous perfusion Intermittent I Intermittent II

Valuesare mean ± standard error of the mean. In each column,valuesthat differ significantly have no letters in common (one-wayanalysis of variance).

beneficial effect is presumed to result from improved coolingof the in situ heart, replenishment of substrate and oxygen, and washout of metabolites produced during ischemia. Similar principles regarding prevention of tissue destruction should apply to long-term preservation scenarios as they do to cardioplegia scenarios, even though there are differences between in situ cardioplegia and in vitro cardiac preservation, particularly the difference in length of arrest time and the presence of noncoronary collateral flow during arrest in situ. The available data indicate that methods and conditions must be carefully selected to ensure success with either short-term or long-term heart preservation. The major criteria of success are generally accepted to be normal recovery of cardiac function at the time of reperfusion, with the ability to support a physiologic workload, and long-term survival of the organ and the host animal. We selected the isolated working rat heart as the experimental model for the present experiments. This type of model has also been used by others I7, 18 to study myocardial protection during cardioplegia and preservation, and it is an important, appropriate initial model for such investigations. In our model, ejecting hearts function at the same workload as in vivo hearts, as determined by myocardial oxygen consumption, pressure indexes, and output indexes, and it thus provides a test of the ability of the preserved organ to function normally in the immediate postpreservation period. Our isolated heart model exhibits remarkable stability of function over prolonged perfusion periods (more than 4 hours). Our data are therefore not confounded by the possibility that the model itself is undergoing progressive ischemia during the perfusion. Favorable results from these small-heart experiments provide a foundation for the study of large animal hearts, including transplantation experiments, and eventually for clinical application. Our data indicate that intermittent perfusion preservation can be used to preserve hearts for extended time periods. The intermittently perfused hearts exhibited postpreservation function that was not significantly different from the function of continuously perfused hearts.

LV+RVdry weight (mg)

181.5 ± 182.5 ± 177.2 ± 180.0 ±

6.2A 8.3A 9.2A 8.3A

Fluid content of hearts (ml waterfgm dry weight)

5.14 ± 5.42 ± 5.85 ± 5.66 ±

0.09A O.13 A, B 0.22B O.loA· B

Valuesare mean ± standard error of the mean. In each column,values that differ significantly haveno letters in common(one-wayanalysisof variance). Hearts that had no perfusionweighed200.9 ± 9.2 mg;their fluidcontent was not included in this analysis because their reperfusion time was not comparable with the times of hearts in the other four groups. LV, Left ventricle;RV, right ventricle.

The intermittently perfused hearts did, however, exhibit significant differences in function compared with fresh control hearts; the results indicated that compliance and contractility were somewhat compromised in the intermittently perfused hearts. The hearts preserved with continuous perfusion exhibited postpreservation function that was not statistically different from that of fresh control hearts, except for dP jdtmax, which was an average of 11% lower in continuous perfusion hearts as compared with control hearts during the 4-hour perfusion period. The relatively small standard deviation of the dP jdtmax data may have contributed to its statistical significance at a relatively small group difference (Table V). Considering all the contractile and output indexes measured, the percentage recovery of function for the hearts preserved with continuous perfusion in the present study ranged from an average of 82.7% recovery of peak power to an average of 92.5% recovery of LVSP. This overall result was better than that which we obtained previously for a continuously perfused group of crystalloid-preserved rat hearts.!? For example, in that study.l? preservation with modified Burt crystalloid solution 19 at P02 545 ± 19 mm Hg produced hearts that had an average of 70% recovery of stroke work. In the present investigation, continuous perfusion preservation with Wicomb's solution!' at P02 820 ± 5 mm Hg produced hearts that had an average of 86% recovery of control heart stroke work. The better overall recovery observed in the present experiments may have resulted from the difference in crystalloid media or preservation P02 in the two experiments or from both of those methodologic differences. Note, however, that the average 82.7% to 92.5% recovery of function of rat hearts preserved with continuous perfusion in the present study is not the maximum recovery that can be achieved with this experimental design (rat hearts preserved for 12 hours at 12° C). In our previousstudy, 10 we obtained 95% to 100%recovery of rat heart function using a novel fluorochemical medium.

The Journal of Thoracic 'and Cardiovascular Surgery November 1993

8 20 Segel and Follette

I

J:.

L.n c:i~ • I

~

Q) ca::::l

....

u

0

..!!!: Ci E E

LAC/SW

40 fo30

..

20 f010 fo0

.--

0.0

0.5

-

1.0

1.5

2.0

I

I

I

2.5

3.0

3.5

~

•

4.0

Hours of Paced Working Heart Perfusion Fig. 5. Rate oflactate production bynonpreserved and preserved isolated working rat heartsduring4 hoursofpaced antegrade perfusion. Lactate production rates (LAC) are shown normalized, based on stroke work (SW) for the 30-minute intervaloverwhich each rate was measured. e, Controlgroup; 0, continuous perfusion group; ., intermittent group II; £1, intermittent group I. Experiments involving intermittent perfusion with fluorochemical media remain to be done. There was scant evidence in this study that the 10-minute perfusion cycle during preservation was more protective of the hearts than was the 5-minute perfusion cycle. The only statistical evidence to that effect was found in the stroke work, lactate production, and tissue fluid content data. Those values for intermittent group I hearts were significantly different from control heart values, whereas values for intermittent group II hearts were not different from control heart values. The two intermittent groups, however, did not differ from each other in any index measured. Indeed, even the stroke work, lactate, and tissue fluid content data do not provide compelling evidence that the hearts of the intermittent groups exhibited differences that would be considered biologically significant. A 5-minute period was apparently sufficient time for washout of metabolic end-products, reestablishment of pH, and resynthesis of energy-rich phosphate compounds, and the additional 5 minutes in each cycle did not have a significant benefit for these hearts. We found evidence, however, that the contractile function of the intermittently perfused hearts differed from that of fresh hearts; that difference was likely the result of the repeated ischemia and reperfusion periods during preservation. The intermittently perfused hearts exhibited a characteristic ischemic metabolic response (i.e., increased lactate production) during preservation. During the 4-hour working heart period, the lactate production/stroke work relationship of the groups suggested that intermittently perfused hearts were also more ischemic after preservation compared with control group and continuous perfusion group hearts. The persistence of

the relatively higher lactate production in the hearts of both intermittent groups during the postpreservation reperfusion phase suggests a higher rate of nonoxidative metabolism in those hearts. Whether this was accompanied by actual tissue damage cannot be established from the present experiments; we did not measure enzyme leakage from these hearts. In the present investigation, we found that cardiac function recorded during the Langendorff perfusion period was not adequate to distinguish group differences that did become apparent when the hearts were in the ejecting mode. For example, all four groups had indistinguishable left ventricular pressure indexes during Langendorff perfusion, yet, during working heart perfusion, significant group differences in contractile indexes were observed (although coronary flow during Langendorff perfusion did suggest the eventual ranking of the four groups). The likely explanation for this observation is that hearts with higher coronary vascular resistance during the Langendorff period were comparatively ischemic and could not maintain a high level of function when oxygen demand was increased during the working heart period. Apparently, the mismatch of oxygen supply and demand in the compromised hearts of both intermittent groups was not great enough during the nonworking mode to cause decreased left ventricular pressure indexes. In a previous study.l'' we noted that nonworking contractile function was suitable as an indicator of long-term preservation efficacy only if the hearts had very poor or no working function. Groups with working function near to that of control, like those in the present study, could not be distinguished during Langendorff perfusion. When the hearts were initially placed on working heart

The Journal of Thoracic and Cardiovascular Surgery Volume 106, Number 5

perfusion, output indexes, as well as contractile indexes, were significantly lower for both intermittent groups compared with the control group. However, the group differences in output resolved after electrical pacing began; during the 4-hour paced perfusion period, group differences occurred only in contractile indexes. The difference in heart rate between the non paced and the paced periods (264 versus 322 beats/min) may have had some influence on this result. Alternatively, the intermittently perfused hearts may have recovered relatively slowly after preservation. There was evidence for the latter possibility: some of the functional indexes (including coronary artery flow and total cardiac output) for both intermittent groups increased during the early part ofthe 4-hour paced perfusion period. A variety of solutions have been tested or used clinically as cardiac preservatives.v II, 19-25 We used the solution of Wi comb and associates II for the present study. In preliminary experiments, we found that coronary occlusion occurred if we used a bicarbonate-buffered medium, such as the solution of Burt and associates.P 19for intermittent perfusion in our preservation apparatus. We believe that the occlusion was caused by CaC03 microcrystals forming in static, nongassed medium that remained in the aortic cannula during the 25-minute no-flow periods and subsequently entered the coronary vasculature. The coronary occlusion could be prevented by using media that did not contain a bicarbonate/carbon dioxide buffer; therefore, for the experiments reported here, we decided to use the phosphate-buffered solution described by Wicomb and associates. II Finally, our experiments support the conclusion that perfusion preservation is superior to simple submersion for long-term hypothermic storage of hearts. Although the length of time that a heart can be successfully stored without adverse consequences may depend on a number of factors, such as species, heart size, temperature, and type of medium, our experiments showed that conditions resulting in failure with submersion-stored hearts produced reasonably well-preserved hearts when perfusion, rather than submersion, was used. We thank Laura Castellanos, BS, Michael Dicus, and Moin Vera for their excellenttechnical assistance. We are grateful to Neil H. Willits, PhD, of the Department of Statistics for assistance with the statistical analyses, Kendall Dempster for advice on the solenoid valve and controller, and Eli Lilly and Co. for the gift of dobutamine HCI. REFERENCES I. Minten 1, Segel LD, VanBelleH, Wijnants 1, Flameng W. Differences in high-energy phosphate catabolism between

Segel and Follette

821

the rat and the dog in a cardiac preservationmodel. 1 Heart Lung Transplant 1991;10:71-8. 2. Spray TL, Watson DC, Roberts We. Morphology of canine hearts after 24 hours' preservation and orthotopic transplantation. 1 THORAC CARDIOVASC SURG 1977;73: 880-6. 3. Wicomb WN, Cooper DKC, Barnard CN. Twenty-fourhour preservation of the pig heart by a portable hypothermic perfusion system. Transplantation 1982;34:246-50. 4. Qayumi AK, lamieson WRE, Rosado LJ, et al. Preservation techniques for heart transplantation: comparison of hypothermic storage and hypothermic perfusion. 1 Heart Lung Transplant 1991;10:518-26. 5. Engelman RM, LevitskyS, eds. A textbook of clinical cardioplegia. Mount Kisco, NY: Futura, 1982. 6. Hearse Dl, Braimbridge MV, lynge P. Protection of the ischemic myocardium: cardioplegia. New York: Raven Press, 1981:353-74. 7. von Oppell UO, Du Toit EF, King LM, et al. St. Thomas' Hospital cardioplegic solution: beneficial effect of glucose and multidose reinfusionsof cardioplegic solution. 1 THORAC CARDIOVASC SURG 1991;102:405-12. 8. Asimakis GK, Sandhu GS, Conti VR, Sordahl LA, Zwischenberger lB. Intermittent ischemia produces a cumulative depletion of mitochondrial adenine nucleotides in the isolated perfused rat heart. Circ Res 1990;66:302-10. 9. Flameng W, DyszkiewiczW, Minten 1. Energy state of the myocardium during long-term coldstorage and subsequent reperfusion. Eur 1 Cardiothorac Surg 1988;2:244-55. 10. Segel LD, Minten lMO, Schweighardt FK. Fluorochemical emulsion APE-LM substantially improves cardiac preservation. Am 1 PhysioI1992;263:H730-9. II. Wicomb WN, Novitzky D, Cooper DKC, Rose AG. Forty-eight hours' hypothermic perfusion storage of pig and baboon hearts. 1 Surg Res 1986;40:276-84. 12. Segel LD, Rendig SY. Isolated workingrat heart perfusion with perfluorochemicalemulsion Fluosol-43. Am 1 Physiol 1982;242:H485-9. 13. Boyle WA, Segel LD. Attenuation of vasopressin-mediated coronary constriction and myocardial depression in the hypoxic heart. Circ Res 1990;66:710-21. 14. Gutmann I, Wahlefeld AW. L-( + j-lactate determination with lactate dehydrogenaseand NAD. In: Bergmeyer HU, ed. Methods of enzymatic analysis. 2nd ed. Deerfield Beach, Fla.: Verlag Chemie, 1974:1464-8. 15. Engelman RM, Auvil 1, O'Donoghue MJ, LevitskyS. The significance of multidose cardioplegia and hypothermia in myocardial preservationduring ischemicarrest. J THORAC CARDIOVASC SURG 1978;75:555-63. 16. Nelson RL, Fey KH, FolletteDM, et al. Intermittent infusion of cardioplegic solution during aortic cross-clamping. Surg Forum 1976;27:241-2. 17. Bernard M, Menasche P, Canioni P, et al. Influence of the pH of cardioplegic solutions on intracellular pH, high-energy phosphates, and postarrest performance. J THORAC CARDIOVASC SURG 1985;90:235-42. 18. Ledingham SJM, Braimbridge MV, Hearse DJ. Improved

8 22 Segel and Follette

myocardial protection by oxygenation of the St. Thomas' Hospital cardioplegic solutions. J THORAC CARDIOVASC SURG 1988;95: 103-11. 19. Burt JM, Larson DF, Copeland JG. Recovery of heart function following 24 hours' preservation and ectopic transplantation. J Heart Transplant 1986;5:298-303. 20. Billingham ME, Baumgartner WA, Watson DC, et al. Distant heart procurement for human transplantation. Circulation 1980;62 (Suppl):IlI-9. 21. Tago M, Subramanian R, Kaye MP. Light and electron microscopic evaluation of canine hearts orthotopically transplanted after 24 hours of extracorporeal preservation. J THORAC CARDIOVASC SURG 1983;86:912-9.

The Journal of Thoracic and Cardiovascular Surgery November 1993

22. Novick WM, Wallace HW, Root KL, Rozanski DJ, Fuller EO. Preservation of donor heart function and high-energy stores by continuous perfusionwith synthetic plasma at 22° C. Circulation 1986;74(Suppl):III80-8. 23. English TA, Foreman J, Gadian DG, Pegg DE, Wheelson D, Williams SR. Three solutions for preservation of the rabbit heart at 0° C. J THORAC CARDIOVASC SURG 1988;96:54-61. 24. Swanson DK, Dufek JH, Kahn DR. Myocardial preservation for transplantation. Transplant Proc 1979;11:1478-9. 25. Kolata RJ, Standeven J, Codd JE. Left ventricularfunction and high-energy phosphates after 24 hours of hypothermic preservation. Heart Transplant 1984;3:319-22.

Bound volumes available to subscribers Bound volumes ofTHE JOURNAL OF THORACIC AND CARDIOVASCULAR SURGERY areavailable tosubscribers (only) forthe 1993 issues from thePublisher, at a cost of $75.00 fordomestic, $98.25 forCanadian, and $93.00 forinternational subscribers forVol. 105 (JanuaryJune) and Vol. 106 (July-December). Shipping charges are included. Each bound volume contains a subject and authorindex and all advertising isremoved. Copies areshipped within 60 days afterpublication ofthelastissue ofthevolume. Thebinding isdurable buckram with theJOURNAL name, volume number, andyearstamped in gold onthe spine. Payment must accompany all orders. Contact MosbyYearBook, Inc.,Subscription Services, 11830 Westline Industrial Drive, S1. Louis, Missouri 63146-3318, USA; phone I (800) 453-4351 or (314) 453-4351. Subscriptions mnst be in force to qualify. Bound volumes are notavailable in place of a regular JOURNAL subscription.